It is well established for both rat and man that the total testosterone concentration within the testes is far higher than that in serum. We know for the that intratesticular testosterone can be reduced by 50-60% without effect on spermatogenesis, but that the required testosterone concentration is still 10-fold greater than serum testosterone concentration. This kind of information, if available for the human, could prove invaluable for understanding and treating at least subsets of men with infertility. Unfortunately, we know little about the androgen content of intratesticular fluid within the human testis or the relationship between intratesticular androgens and human spermatogenesis. Our recent studies of the human have demonstrated that, as in the rat, there is a gradient between the concentration of testosterone in serum and within the testis; intratesticular testosterone levels were found to be 100-fold higher than normal serum testosterone levels. We do not yet know how much of the testosterone within the human testis is required either to maintain or restore quantitatively normal spermatogenesis because, as yet, experimental studies comparable to those performed in the rat have not been feasible for the human. Moreover, we know little about the concentration of bioavailable androgens within the testes of any mammal, and therefore virtually nothing about the relationship between bioavailable androgen concentration and spermatogenesis in rat or man. The major objectives of this project are to identify and quantify the androgen content of the human testis, to assess the bioavailability of intratesticular androgens, to experimentally determine the relationship between bioavailable intratesticular androgen concentration and spermatogenesis, and to examine the effect of testosterone replacement modalities on intratesticular bioavailable androgen concentration and on spermatogenesis in subfertile men.